paleotree (version 2.5)

simPaleoTrees: Simulating Un-Conditioned Trees of Fossil Taxa

Description

Creates sets of paleo-trees with minimal conditioning and with sampling such that lineages may be missing and terminal branches are truncated, but node-times are known perfectly

Usage

simPaleoTrees(p, q, r, ntrees = 1, all.extinct = FALSE,
  modern.samp.prob = 1, mintime = 1, maxtime = 100, mintaxa = 2,
  maxtaxa = 500, anag.rate = 0, prop.bifurc = 0, prop.cryptic = 0,
  drop.zlb = TRUE, print.runs = FALSE, ranges.only = TRUE, plot = FALSE)

Arguments

p

Instantaneous rate of origination/branching per lineage-time units.

q

Instantaneous rate of extinction per lineage-time units.

r

Instantaneous rate of sampling per lineage-time units.

ntrees

Number of trees to simulate

all.extinct

Condition on all taxa being extinct by modern? Default is false

modern.samp.prob

Probability of sampling living taxa at the present day (time=0), see documentation for sampleRanges

mintime

Minimum time units to run any given simulation before stopping.

maxtime

Maximum time units to run any given simulation before stopping.

mintaxa

Minimum number of total taxa over the entire history of a clade necessary for a dataset to be accepted.

maxtaxa

Maximum number of total taxa over the entire history of a clade necessary for a dataset to be accepted.

anag.rate

Instantaneous rate of anagenesis (i.e. pseudospeciation/pseudoextinction).

prop.bifurc

Proportion of morphological branching by bifurcating cladogenesis relative to budding cladogenesis.

prop.cryptic

Proportion of branching events with no morphological differentiation (i.e. cryptic speciation) relative to branching events associated with morphological differentiation (budding and/or bifurcating cladogenesis). Trees with cryptic taxa are always retu

drop.zlb

Should zero-length branches be dropped?

print.runs

If TRUE, prints the proportion of simulations accepted for output to the terminal.

ranges.only

If TRUE (the default), the ranges returned in $ranges are given as taxon first and last occurrences only. If FALSE, $ranges returns the times of all sampling events for each taxon as vectors within a list.

plot

Should data be plotted as it is simulated?

Value

• Output is an object of class multiphylo containing the simulated phylogenies, unless ntrees is one in which case the output is a phylogeny of class 'phylo'. Additionally, each of these simulated phylogenies will have the original simulated taxa data (from simFossilTaxa) and sampled ranges (from sampleRanges) attached as the elements $taxa and $ranges to each phylo object.

Details

This function is essentially a complex wrapper for simFossilTaxa, sampleRanges and taxa2phylo, and users should refer to the respective documentation for those functions in addition to this manual page: simFossilTaxa, taxa2phylo and sampleRanges. This function will output simulated phylogenies of fossil taxa where the divergence times are perfectly known, only sampled taxa are included on the tree and tips are located at the last observed time for the species (the apparent time of extinction, except for living taxa). simPaleoTrees essentially uses simFossilTaxa with no.cond as TRUE and other minimal conditioning, so as to get as unbiased a sample of simulations as possible (without exceeding the maximum constraints). This is useful for birth-death analyses, although a number of options available in simFossilTaxa are thus unavailable in simPaleoTrees. By default, there is no conditioning on the number of extant taxa, living taxa are sampled perfectly at time 0 and zero-length branches are dropped. Unlike simFossilTaxa, you cannot condition on a certain number of extant taxa, only whether they are allowed or not (via all.extinct). As of version 1.6, there are now options relating to speciation modes. By default, taxa are only simulated under budding cladogenesis but this can be changed with the arguments anag.rate, prop.bifurc and prop.cryptic. Trees with cryptic taxa are always returned with cryptic taxa unmerged (see sampleRanges). Because the divergence times are known perfectly, yet tips are at the apparent time of extinction and unsampled taxa are dropped, one should not use the output of this analysis except for very specialized simulation analyses. The results are probably not anything like real datasets of paleontological phylogenies, at least in most aspects. The print.runs argument does not work precisely as in simFossilTaxa; it only counts how many accepted datasets from simFossilTaxa are acceptable for output after the simulation of sampling.

See Also

simFossilTaxa, taxa2phylo, sampleRanges

Examples

set.seed(444)
#simulate trees conditioned to have no living descendants
trees <- simPaleoTrees(p=0.1,q=0.1,r=0.5,ntrees=10,all.extinct=TRUE,maxtime=100,
    print.runs=TRUE,plot=TRUE)
#number of tips
sapply(trees,Ntip)

#simulate trees conditioned to have possible living taxa and perfect sampling at modern
trees <- simPaleoTrees(p=0.1,q=0.1,r=0.5,ntrees=10,all.extinct=FALSE,maxtime=100,
    modern.samp.prob=TRUE,print.runs=TRUE,plot=TRUE)
#number of tips
sapply(trees,Ntip)

#simulate trees with no living descendants and return sampling events, not ranges
trees <- simPaleoTrees(p=0.1,q=0.1,r=0.5,ntrees=10,all.extinct=TRUE,maxtime=100,
    print.runs=TRUE,plot=TRUE,ranges.only=FALSE)
#number of sampling events
sapply(trees,function(x) sum(sapply(x$ranges[!is.na(x$ranges)],length)))
#remove any sampling events at t=0; shouldn't be any, so should be identical
sapply(trees,function(x) sum(sapply(x$ranges[!is.na(x$ranges)],function(y) sum(y>0))))